METHOD AND SYSTEM FOR DETERMINING RATES OF INSURANCE
The invention provides a method and system for determining rates of insurance for travelers based on certain risk parameters. Insurance rates are computed using factor relativities derived from insured data and parameters for several different insurance coverages. Factor relativities are computed through an iterative process using a minimum bias procedure according to each coverage. Parameters are evaluated simultaneously for each coverage to eliminate cross bias from one parameter to another. Once factor relativities are calculated, the system and method allows for a consumer to request a rate on an insurance coverage. The system and method provides the rate using the factor relativities for the requested coverage. The method and system also provides insurance rates to additional or upgraded insurance coverages based on consumer preferences.
Latest American International Group, Inc. Patents:
- GENERATING TREND DATA FOR A CYBERSECURITY RISK SCORE
- System and method for assessing cybersecurity risk of computer network
- Systems and methods for dynamic real-time analysis from multi-modal data fusion for contextual risk identification
- Scrubbing member resistance
- SYSTEM AND METHOD FOR ASSESSING CYBERSECURITY RISK OF COMPUTER NETWORK
This patent application claims the benefit of U.S. Provisional Patent Application No. 60/956,865, filed Aug. 20, 2007, which is incorporated by reference.
FIELD OF THE INVENTIONThe invention generally relates to a method and system for determining rates for insurance. More particularly, the invention relates to a method and system for determining base insurance rates using multiple risk parameters and factors.
SUMMARY OF THE INVENTIONInsurance rates are computed using factors derived from insured data and parameters for several different insurance coverages. Factors are computed through an iterative process using a minimum bias procedure according to each coverage. Parameters are evaluated simultaneously for each coverage to eliminate cross bias from one parameter to another. Once factor relativities are calculated, the system and method calculates a rate on an insurance coverage using the factors for the requested coverage. The method and system also provides insurance rates for additional or upgraded insurance coverages based on consumer preferences. The underwriting methodology as described in detail below provides unexpected results. By isolating individual factors and eliminating cross biases, a more accurate rate may be established.
Embodiments of the present invention provide a method and system to offer insurance at competitive rates and custom coverage packages. Embodiments of the present invention use multi-factor based pricing and a recursive underwriting procedure to determine the insurance rates and custom coverage packages. The inventive method and system may be applied to a variety of insurance products, including particularly travel insurance. Although the following description explains various features of the invention as related to travel insurance, persons of skill in the art will appreciate that the described underwriting methodology and systems may be applied to other types of insurance.
From an actuarial perspective, factor based pricing permits an insurance product to be rated through the use of multiple factors, each evaluated simultaneously with all other parameters. This underwriting approach differs from others that do not evaluate the effect of one parameter on another. The final determination of rates is through the calculation of factor relativities that are determined through an innovative recursive and simultaneous evaluation process.
The factors for each parameter are determined for each coverage within a travel insurance package. Factor relativities are derived from a minimum bias approach to apply on a multiplicative basis to generate an appropriate price for the insurance package. Each coverage has a base premium, and depending on the characteristics of the insured, factors relating to those characteristics are multiplied successively against the base premium. The factors are called “relativities” because they generally relate to the proportional differences between a predetermined set of individual characteristics and the selected set.
The travel insurance company 116 implements a web server 118 that gathers consumer information and displays different insurance products and rates through the travel insurance company's website. A processor 124 calculates the factor relativities using a minimum bias approach. A data warehouse 122, which may be a collection of databases, stores high quality actuarial and insurance data that is used to determine factor relativities. A ratings engine 120 accesses the calculated factor relativities from the data warehouse 122 to thereafter calculate a rate of an insurance product requested by a consumer 114 or travel agent 106.
At step 200, a query analyzer queries a database to access insured data. The analyzer organizes the data into bands for each parameter and losses by coverage for each set of parameter band possibilities. The query analyzer thereafter converts the data into a flat file, step 204, to permit the data to be transferred into a smaller database, step 204, for further analysis. The step of transferring data extracted from a larger database into a smaller one facilitates faster analysis of the extracted data. In the preferred embodiment illustrated in
At step 206, an application 238 accesses the data table from the Access database 236. The application 238 places seed factors and a pure premium base for each coverage for all parameter bands into arrays, step 208. Seed factors are the initial inputs used to produce the final factor relativities that, when applied to a base premium rate, are used to calculate a final insurance rate. The seed factors are determined by evaluating the insurance package results stored in the data warehouse 122 for each parameter. The pure premiums (loss costs) are calculated for each of the exemplary bands above. A certain band (e.g. the most populous band) is defined as the base band and all remaining bands are divided by this band's loss cost. The base band is therefore set to 1.000 and other bands are stated relative to the base band. The base premium amount is adjusted in the opposite fashion to counteract the impact of subsequent adjustments. Exemplary seed factors for certain parameters are shown in
Seed factors are not the final factors because each seed factor is determined without regard to how each rating band is impacted by other factors. The seed factors are a reasonable starting point for a series of recursive calculations that incrementally adjust the factor. Through this iterative process, the factor converges on a value that accurately reflects the expected and isolated loss associated with the factor. Seed factors typically contain considerable cross bias when compared with the final factor relativities derived from the minimum bias procedure. The factor relativities demonstrate that the relationship by coverage with each parameter varies more than the seed factors. Determination of final factors through a recursive bias elimination process enables a more accurate rating, particularly as different coverage limits are proposed, than conventional approaches to travel insurance underwriting.
Steps 210 through 216 of
At step 222, application 238 exports the factors into a spreadsheet 240. At step 224, the factor relativities are reviewed. Decisional 226 indicates that if the parameter band factors do not logically relate to actual loss experience, the process must start anew. However, if the parameter band factors are logically related to actual loss experience, the factor relativities will be manually adjusted, as needed, to account for outliers, step 228. Next, at step 230, the coverage pure premium base is divided by the target loss ratio to obtain a final rate factor.
At step 306, a consumer or travel agent requests the rate of the product. This request may be automatically prompted by the travel agent program or by an independent and manual process. At a next step 308, the ratings engine validates business rules on the entered data. The ratings engine thereafter determines whether the business rules are validated, step 310. An example of consumer data invalidating a business rule may be entering a destination that has no parameter band or entering an invalid Trip Type. If any of the consumer data violates a business rule, then the user interface prompts the consumer or travel agent to re-enter the data 304. However, if the business rules are validated, then the ratings engine determines whether the rating is by benefit. If so, then the ratings engine continues to a next step 314. If not, then the ratings engine continues to a next stage 316.
At step 314, the ratings engine accesses the benefits from a SQL database server. Exemplary lists of benefits are shown in
At a next step 318, a ratings engine finds the rating factor and rating rule. The rating factor may be the premium base rates as shown in
A factor based pricing model using the minimum-bias procedure (where
Not all coverages are affected by the parameter selections in the same way. Embodiments of the invention incorporate separate parameters for each coverage by the parameter bands providing a more accurate rating and greater flexibility in adding additional coverage specific upgrades to the product. This approach also allows for easier adjustment of benefit levels of the coverages. Embodiments of the invention contain additional programming to adjust for simultaneous evaluation of factors on multiple coverages. Additional arrays are added to the calculation program as well as adding more columns to the matrices. Also, the factor adjustment computation part of the program is altered to accommodate multiple coverages. Embodiments of the invention implement this by adding additional columns (two for each coverage) to the one-dimensional parameter specific matrix. The first set of columns for each coverage holds the actual losses. The second set of columns holds the calculated losses using the coverage pure premium base on the currently stored factors in the multiple arrays (if this is the first iteration then the arrays hold the seed factors).
The following example uses the data in
There are several examples of optional insurance coverages. The minimum bias approach allows embodiments of the invention to offer rates on optional packages by the same or different relativity factors, as necessary, and track the premium separately by coverage.
The minimum bias approach as illustrated in
Optional packages may include increasing coverage limits on medical expenses and baggage, and also selecting medical coverage to be primary medical insurance from secondary medical insurance. These packages are considered together when calculating the factor relativities using the previously described underwriting methodologies. Thus, factor relativities are computed for each separate package for each option. Then the rate of the package is computed by the ratings engine.
Other optional packages may include Increased Evacuation, Bring Them Home, Added Perils, and Adventure Sports Rider. Added Perils is coverage for Default, Terrorism and Pre-Ex if purchased within 15 days of deposit is an option that is packaged together. As with other optional and upgrade coverages, the previously adjusted amounts are added, and applied to the appropriate coverages and rating factors.
Embodiments of the invention allow for the coverage and limits to be accessed and distributed in several ways. Consumers may purchase an insurance product from a website. A consumer's state of residence may limit the insurance products available them to only those coverages where the insurance product is approved. Further, consumers may also purchase the insurance product through a call center. In addition, travel agents are able to offer and sell insurance products to their clients via a website. Products may also be marketed through online aggregators.
The following example an exemplary method for the iterative steps 206-220 in
At an exemplary step 206, a software application accesses data, as shown in Table 1, from a database. The data in Table 1 contains the Actual Losses for a Trip Cancellation (TC) and a Trip Interruption (TI) benefit organized into Trip Cost and Age parameter bands.
As part of exemplary step 208, a software application calculates the Base Loss Costs for TC and TI. The Base Loss Cost for a benefit is calculated by dividing the sum of the losses for each parameter band for the benefit by the sum of the insureds. The TC Base Loss Cost and TI Base Loss Cost are shown in Table 2.
Also as part of exemplary step 208, seed factors are selected for the iterative procedure. Seed factors are not the final factors because each seed factor is determined without regard to how each rating band is impacted by other factors. As previously noted, the seed factors are a reasonable starting point for a series of recursive or iterative calculations that incrementally adjust the factor. The software application applies an iterative and recursive calculations to converge on a factor value that accurately reflects the expected and isolated loss associated with the factor. A software application implements the iterative procedure due the complexity and high number of calculations that are needed to be performed. The seed factors for this example are shown in Tables 3 and 4.
As part of the iterative procedure, any iterative minimizing procedure may be used to find the relative factors for a parameter. These include a Balance Principle, Least Squares, or Chi-Squared approach. The general form of each equation is shown as follows:
Balance Principle Equation=xi=Σnijrij/Σ . . . Σnijyj . . . zk,
Least Squares Equation: xi=Σ . . . nij2rijyj . . . zk/Σ . . . Σnij2yj2 . . . zk2
Chi-Squared Equation: xi={Σ . . . Σ[nijrij2/(yj . . . zk)]/Σ . . . Σnijyj . . . zk}0.5
where xi=Relative Factor for the Parameter, n=number of insureds, r=actual loss cost, y . . . z=Factors for each parameter y . . . z, and i, j, . . . k represent the number of bands for each factor x, y, . . . z, respectively.
In this example, the Balance Principle, Least Squares, and Chi-Squared formulas for the two parameters (Trip Cost and Age) are listed below:
Balance Principle Equation=xi=Σnijrij/nijyj,
Least Squares Equation: xi=Σnij2rijyj/Σnij2yj2
Chi-Squared Equation: xi=[Σ(nijrij2/yj)/Σnijyj]0.5
where x=Trip Cost Factors, n=insureds, r=actual loss cost, y=age factors, and i and j represent the number of bands for a trip cost factor and an age factor, respectively.
The final factor relativities for the Trip Cost and Age parameters can be found using the iterative procedure shown in the exemplary steps 210-220 in
The iterative procedure described above minimizes cross-biases between factors to accurately price insurance products with respect to loss. Pricing mechanisms used in the prior art for travel insurance products typically provide a loss ratio of 0.42. However, experience has shown rates found using the invention for the same travel insurance products to improve the loss ratio substantially, e.g., 0.31. The invention results in a 25% improvement, or more, over the prior art, thereby providing a more accurate pricing of insurance products.
The use of the terms “a” and “an” and “the” and similar referents in the context of describing the invention (especially in the context of the following claims) are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. The terms “comprising,” “having,” “including,” and “containing” are to be construed as open-ended terms (i.e., meaning “including, but not limited to,”) unless otherwise noted. Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicated herein, and each separate value is incorporated into the specification as if it were individually recited herein. All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g., “such as”) provided herein, is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the invention.
Preferred embodiments of this invention are described herein, including the best mode known to the inventors for carrying out the invention. Variations of those preferred embodiments may become apparent to those of ordinary skill in the art upon reading the foregoing description. The inventors expect skilled artisans to employ such variations as appropriate, and the inventors intend for the invention to be practiced otherwise than as specifically described herein. Accordingly, this invention includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the invention unless otherwise indicated herein or otherwise clearly contradicted by context.
Claims
1. A method for determining a rate of insurance, the steps comprising:
- analyzing insured data to determine parameters and parameter bands needed for a plurality of insurance benefits;
- storing parameter and parameter band data for the plurality of insurance benefits in a first database;
- accessing parameter and parameter band data for the plurality of insurance benefits from the first database;
- calculating with a computer seed factors based on parameter, parameter band, and premium base rate, for the plurality of insurance benefits;
- calculating factor relativities using an iterative procedure from seed factors, for the plurality of insurance benefits; and
- determining a rate of insurance by applying the calculated factor relativities to a selected insurance product.
2. The method according to claim 1, further comprising the step of providing rates for insurance benefits based on consumer preferences.
3. The method according to claim 1, further comprising the steps of extracting insured data from the first database; and storing the insured data into parameter bands as flat file in a second database.
4. The method according to claim 1, wherein the iterative procedure is selected from the group consisting of a Balance Principle equation, Least Squares Method equation, and Chi-Squared Method equation.
5. The method according to claim 1, wherein one or more software applications that are implemented by the computer and databases are selected from the group consisting of a query analyzer, database program, visual basic program, and a spreadsheet application.
6. The method according to claim 1, further comprising the steps of:
- determining whether a first factor of a parameter in a first iteration is within a threshold of a second factor of the parameter in a second iteration;
- designating the first factor as a final factor when the first factor in the first iteration is within the threshold of the second factor of the second iteration, wherein the final factor is used to calculate the rate of insurance;
- performing another iteration of the iterative procedure when the first factor of the first iteration is not within the threshold of the second factor of the second iteration.
7. The method according to claim 1, wherein the insurance benefits are selected form the group consisting of Trip Cancellation Insurance, Trip Interruption Insurance, Medical Coverage, Evacuation Assistance, Baggage Delay, Increased Evacuation, Bring Them Home, Added Perils, and Adventure Sports Rider.
8. A method for determining a rate of insurance, the steps comprising:
- computing the parameters needed to calculate an insurance rate for a selected insurance product;
- requesting consumer information from a consumer to calculate the insurance rate;
- validating consumer information based on business rules;
- calculating, with a computer, the rate of the insurance product based on independently calculated factor relativities and a premium base rate for each insurance benefit of the insurance product; and
- displaying the rate of the selected insurance product to the consumer.
9. The method according to claim 8, further comprising the steps of calculating the rate of each insurance benefit of the insurance product using an iterative procedure.
10. The method according to claim 8, wherein the iterative procedure is selected from the group consisting of a Balance Principle equation, Least Squares Method equation, and Chi-Squared Method equation.
11. The method according to claim 8, wherein one or more software applications that are implemented by the computer and databases are selected from the group consisting of a query analyzer, database program, visual basic program, and a spreadsheet application.
12. The method according to claim 8, wherein the parameters are selected from the group consisting of Age, Trip Cost, Trip Length, Destination, State of Residence, Days to Departure, Departure Month, and Trip Type.
13. The method according to claim 8, further comprising the step of determining a rating factor and rating rule for the rate of the insurance product.
14. A system for determining a rate of insurance, the components comprising of:
- a user interface capable of permitting a consumer to request a rate of an insurance product from an insurance carrier;
- a web server that displays insurance products, rate information, and other insurance data to a consumer;
- a storage device containing independent factor relativities calculated using a minimum bias procedure;
- a ratings engine that calculates the rates of insurance products using factor relativities.
15. The system according to claim 14, the components further comprising a software application that calculates the independent factor relativities.
16. The system according to claim 14, the components further comprising a calculation module that accesses the calculated independent factor relativities from the storage device to calculate the rate of the insurance product.
17. The system according to claim 14, the components further comprising of a first database that contains insured data to be used to calculate the rate of the insurance product;
- a second database that contains a converted flat file of the insured data placed into parameter bands, to be used to calculate the rate of insurance.
18. The system according to claim 14, wherein the minimum bias procedure is selected from the group consisting of a Balance Principle equation, Least Squares Method equation, and the Chi-Squared Method equation.
19. The system according to claim 14, wherein one or more software applications that are implemented by the calculation module and the databases are selected from the group consisting of a query analyzer, database program, visual basic program, and a spreadsheet application.
Type: Application
Filed: Aug 20, 2008
Publication Date: Feb 26, 2009
Applicant: American International Group, Inc. (New York, NY)
Inventors: Joe TRITZ (Custer, WI), Kristina Smith (Stevens Point, WI), Jason Schilling (Berlin, WI)
Application Number: 12/195,231
International Classification: G06Q 40/00 (20060101); G06F 17/00 (20060101); G06Q 90/00 (20060101);